Public Polls

Credibility of public polls has been a bone of contention recently after June 2 local elections.

The controversy flared up as the election results were quite different from pre-election forecasts based on opinion polls revealed by media.

The error was much greater than margins commonly permissible in social sciences. The wrong predictions are largely due to the change of life-styles and communication method in the era of mobile communication. Voices are loudly heard, calling for change of polling method and media¡¯s manipulation of opinions based on incorrect surveying.

Public polls are the most common method to survey the facts, trends, and opinions of the people mostly by written forms, telephones, emails, Internet, social networks and broadcasters. Whatever means a poll may adopt to survey, the result may have potential for inaccuracy due to sampling error which reflects the effects of chance and uncertainty in the sampling process.

Some people suggest that a poll with a random sample of 1,000 people has a margin of sampling error of 3 percent for the estimated percentage of the whole population. This margin of error can be reduced by using a large sample. If a pollster wishes to reduce the margin of error to 1 percent, they say a sample of around 10,000 people is necessary. However, a survey on large number of sample incurs more expense. Therefore, pollsters are inclined to balance the cost of a large sample against the reduction in sampling error and a sample size of around 500-1,000 is a typical compromise for political polls.

According to the difference in propensity, location, education level, living standard, religion, and cultural or political background of the people surveyed, the results may differ remarkably. And as situation and environment changes, the opinion of the surveyed people can change also, making the surveyed result erroneous.

For example, in political campaign polls, some people would not reveal their true opinion at the survey, as seen a theory of ¡°spiral of silence.¡± But who knows and who can prove that the people would cast the ballots as they answered at the time of public survey?

Therefore, no one can guarantee the survey result would be true or false in the end. However, generally speaking, the credibility of the survey results rises as the number of samples is increased. Usually, the number of samples, sampling method, survey process, analysis method, and statistical reliability of the survey result are major ingredients which influence the survey results. So, would you believe the result of the survey or not?

Opinions are diverse on whether or not polls are continuously conducted and how they are done in the era of mobile communication. Those in support of continuing polls are of the view that public polls are necessary to diagnose the opinions of citizens, especially voters¡¯ thinking around election days, even though there may be some errors in surveying. They argue that polls can help people judge the social tendency and political circumstances. Others take issue with the side effects of polls, fearing that incorrect results can mislead the public to a wrong direction in the end.